Q1: In the introduction of new TLDs, what steps should
be taken to coordinate with the
Internet Engineering
Task Force, the Internet Architecture Board, and other organizations
dealing with Internet protocols and standards?BT considers that
ICANN should seek the guidance of the IETF, the IAB and other organisations concerned
with the maintenance of the stability of the Root Server system and the network to
determine the capacity of the systems for sustaining increasing traffic loads as
might arise from increases in the numbers of TLDs. Other issues for liaison are related
to:-
· the inefficiency of look-up (probability of forming a bad name, reduced
cache hits);
· the implications of changing DNS field lengths, order, structure;
·
security/performance threats imposed by changing the DNS topology and topography
·
the guidance / standardisation of the use of particular new TLD and other non performance
issues.
Q2: What stability concerns are associated with the initial phases of registration
within the TLD?
BT suggests that the implications arising from the demands of
an increasing number of registries on the Root Server system should be carefully
considered when determining how many domains should be added and in what manner and
over what time-scale. The role out of the first new TLD should be completed in a
controlled manner giving users confidence in the new TLDs.
Policy and operational
issues, however, may pose as great a risk to Internet stability as technical ones.
New TLDs that confuse users, reduce the ability to locate pages on the World Wide
Web, or dramatically increase the number of bad faith domain name registrations,
would disrupt the stability of the Internet and/or cause increased legal problems
for those using it.
There is almost bound to be a landrush for new domains when
a TLD is released. This will undoubtedly cause Operational Service Support issues
and the TLD registry(ies) should have an agreed policy for dealing with this.
From an operational perspective, some damage could be done by a new TLD registry
operator that was unable to handle demand for new TLDs, that erred in entering or
processing new registrations, that proved unwilling to operate by ICANN's policies
and procedures, or that was unable to build a viable and lasting operation.
While
the ISP can do this without any of the checks stated to be necessary for a new TLD
registrar simply by gaining a name subspace through normal registration. However
this is changing to more direct registrations (recent spate of national TV adverts
for registering your own domains). If there are barriers to TLDs which greater than
those for subspaces then users will be driven to use subspaces. The sub-domain
control issue is always tricky as once anyone delegates a domain they lose control
of it by definition [details of which are too involved to include here]. It would
be "better" to use subspaces otherwise forget DNS and use a flat file.
Normal business
rules (quality of service, cost etc) will drive out bad registrars. Normal business
rules will be applicable in the future but every effort should be made to ensure
that these first few TLD are a success. Name changes are generally a bad thing and
should be discouraged [for a start they cause increased bad lookups]. The whole point
of a name is that it doesn't change. However it is possible that the addition of
new domains may result in an increasing level of name-changes and moves which means
that such changes will need to be made as easy as possible.
Q3: What can be
done to eliminate or reduce these stability concerns?
New TLDs should be adopted
in accord with the principles recommended by Working Group C and discussed in more
detail below. Domain name registrations in new TLDs should be subject to the current
and all future versions of the UDRP.
Operators of new TLD registries should be
subject to rigorous review to ensure their ability to handle the volume expected
from new registrations, to operate the registry in a technically sound manner, and
to maintain a viable and lasting operation.
Operators of new TLD registries should
maintain copies of all data, software and other code in an escrow that could be accessed
by ICANN in the event that the registry operator needed to be replaced as the operator
of a given TLD. Such escrowed data should be maintained in a manner in which it could
be easily understood and used by a new registry operator. ICANN should implement
such an escrow scheme in a manner designed to ensure the seamless transfer from one
registry operator to another.
BT proposes that one method for ensuring that risks
to the stability of the Internet would be to undertake a trial of a few TLDs and
their registry operators whose impact can be monitored and assessed.
Q4: Would
these stability concerns be magnified by introducing a large number of TLDs
at once?
BT strongly believes that this is likely to be the
case. Given the need to thoroughly review the operational readiness of the operator
for each new TLD registry and the need to ensure that registration policies and procedures
for each new TLD are sound and reduce the stability concerns noted above, given ICANN's
limited resources, British Telecommunications supports the view of the Business and
Commercial Constituency that the concerns noted likely would be magnified by introducing
a large number of TLDs at once.
Q5: Are there any practical means of reversing
the introduction of a significant new TLD
once it
goes into operation?
Reversing the introduction of a permanently-assigned new TLD
is not advisable. BT supports the view of the Business & Commercial constituency
that ICANN should, however, ensure that it is able to transfer the operation of a
TLD from one registry operator to another easily and with minimal disruption in the
event that a registry operator is unable or unwilling to continue performing its
duties.
Q6: Is it feasible to introduce a TLD on a "trial basis," giving clear
notice that the TLD
might be discontinued after the
trial is completed?
BT considers that there is no technical reason why an experimental
TLD cannot be withdrawn at the end of a trial period. However it would be advisable
to make it clear to registrars and registrants that the purpose of such an 'trial'
entity has only limited lifetime and will be withdrawn at the end of the trial, and
a date specified.
However it is difficult to see how a new TLD could be run on
a trial basis, as it might not attract the volumes of registrations necessary to
test-pilot a registry nor the traffic that might be needed to establish its impact
on the stability of the system. If the new TLD is assigned a particular functional
purpose, with no conditional qualifications, as a result of which it is adopted by
registrants as a permanent facility, then it should not be withdrawn.
Given
the significant expense of building and promoting an Internet presence, the need
to ensure that users can find what they are looking for on the Internet, and the
potential financial damage that would be caused to businesses from losing their Domain
Name, it is not practical to think about removing a TLD once it has been established.
Q7:
To ensure continued stability, what characteristics should be sought in a proposed
TLD and in the organization(s) proposing to sponsor and/or operate
it?
BT supports the principles that have been clearly established in the paper
submitted by the Business and Commercial constituency of the DNSO, for selecting
the new TLDs. So long as a new TLD meets each of the seven principles recommended
by Working Group C, it should be acceptable. Once a TLD has met those principles,
the central guide for whether it should be adopted may well be the operational stability
of the proposed TLD registry operator. These are summarised as follows:
The BC
proposes a methodology to create new Internet domain names in a way which simultaneously
meets the global desire for more names and reduces the potential for consumer fraud
and confusion.
The methodology for introducing new names is:
- establish a set
of principles for choosing the names
- write clear guidelines based on those principles
-
empower a panel under the auspices of the Names Council to receive TLD applications
and follow the guidelines
- invite applicant Registries to apply for new TLDs
of their choosing
- accept or reject the applications based on compliance with
the guidelines.
The principles upon which guidelines should be established are:
1.
Certainty: a TLD should give the net user confidence that it stands for what it purports
to stand for.
2. Honesty - a TLD should avoid increasing opportunities for malicious
or criminal elements who wish to defraud net users.
3. Differentiation - the selection
of a TLD string should not confuse net users and so TLDs should be clearly differentiated
by the string and/or by the marketing or functionality associated with the string.
4.
Competition - new TLDs should foster competition in the supply of domain names such
that the authorisation process for new gTLDs should not be used as a means of protecting
existing service providers from competition.
5. Diversity - new TLDs should become
available to meet the needs of an expanding Internet community. They should serve
both commercial and non-commercial goals.
6. Meaning - a gTLD should have meaning
and its selection should include how the new TLD will be perceived by the relevant
population of net users.
7. Simplicity - adherence of the above principles should
not impose an overly bureaucratic procedure on a registry.
BT supports the Business
& Commercial constituency position that a possible long-term solution is to extend
the classification role of TLDs to their logical conclusion. One option is to create
a large set of TLDs to reflect the needs of the consumer. These TLDs are called charter
or topic TLDs. Thus, for example, the following three commercial and one non-commercial
options could all co-exist: Lotus.software Lotus.cars Lotus.paper Lotus.flowers
BT
considers that the selection of a TLD registry operator should be based on its technical
abilities, its willingness to abide by ICANN's policies and procedures, its financial
stability, scalability, the qualifications of management, and the likelihood that
it will be able to provide ongoing services over a long period of time. However BT
also considers that ICANN should provide the scope for a TLD applicant to contract
out the registry operations to an appropriately qualified entity.
Q8: To what extent
is the experience gained from introducing gTLDs in the 1980s applicable to present-day
circumstances?
Scarcity and monopoly do create commercial and service risks which
provide opportunities for abuse. In the situation where the customer understanding
is still low there remain risks which will bring to bear the same land-grab, and
customer confusion issues with DNS that have occurred previously, unless these are
managed in a careful manner, by ensuring that relevant protection measures exist
in all domains.
Q9: To the extent it is applicable, what are the lessons to be
learned from that experience?
BT considers that the reported 'exhaustion' of existing
TLD .com has arisen in a premature way because there has been no sub-domain structure
that allows an efficient use of the available address space. Therefore ICANN
should consider preferentially proposals which include efficient structures for utilising
the sub-domain space.
Q10: What lessons, if any, can be learned regarding new
gTLD introductions from the experience of the ccTLD registries?
It is clear that
the ccTLDs have been a considerable success The major lesson argues for a clear charter
of the use of the TLDs and the power to reassign registries if that charter is flouted.
It is important that ICANN carefully considers the impact of new TLDs on users should
they become a tradable commodity. All TLD rights should be held by, and returned
to ICANN, after lease.
Q11: Can lessons relevant to introduction of new TLDs be
learned from the recent decisions by a number of them to operate in a globally open
manner? If so, what lessons?
ICANN should limit the scope of any new TLD strictly
so it cannot turn into an unforeseen competitor to an existing TLD in an unmanaged
way otherwise this will give rise to confusion for users. This should still allow
for differentiation and competition between alternative domains.
Q12: Is the Names
Council's recommendation that a "limited number of new top-level domains be introduced
initially" a sensible way to minimise risks to Internet stability?
BT supports
the Names' Council's proposal for the introduction a few TLDs followed by a period
of monitoring and assessment. This should enable any adverse impacts to be identified
and rectified before there are serious repercussions for users of the Internet. Factors
which need to be considered here are the range of TLD types which could be sanctioned.
Clearly new TLDs which only represent minority interests may not place significant
strain on respective registries or traffic loading on systems servers. Consequently
the range of TLD types and their rate of introduction should follow an incremental
path, which will enable stepwise evaluation of the increased demands. In addition
inefficient domain structures may cause excessive demands on the infrastructure.
The
uncertainties which need to be used as key measures of impact include:-
· rate
of registrations;
· reduced look-up efficiency if bad TLDs are introduced;
·
DNS cache efficiency;
· theproportion of badly formed URL's
· unforeseen factors
affecting DNS resolution e.g. people adding search lists to their DNS resolver.
Q13:
What steps should be taken to evaluate carefully the initial introduction of TLDs
before future introduction of additional TLDs?
BT considers that the impact of
increasing registries and registrants on the Root Server system and the traffic demands
need to be continuously monitored and evaluated. In addition the success of new TLD
operators in establishing a competitive and differentiable position should be considered,
to avoid the creation of virtual monopolies in specific functional or commercial
areas. This may impose the need to allow competing entrants timely entry as TLD operators.
Q14:
Should a fixed time be established for all the evaluations, or should the time allowed
vary depending on the nature of the TLD and other circumstances?
If a new TLD is
introduced on a trial basis, without any declared intention that there is a finite
limit to the duration of the trial, then it will be unclear to potential registrars
and registrants whether this is experimental or the initial assessment of a permanent
facility. The essential principle must be that any trial should have a clearly defined
end date, and the future intention, termination or extension made clear, even if
the results from the trial require some modification to the operation of the given
TLD or supporting systems.
Under these circumstances the ideal solution is that
any trial should have a finite duration, and that such trials might be restricted
to given classes of TLDs. All assignments should continue to be subject to ICANN
governance and might have to sustain temporary or permanent suspension of new registrations
in the event of operational problems arising at a later date, for reasons that are
not foreseen at the time of creation.
For non- trial introductions there probably
should be a fixed time-scale for the evaluation.
Q15: Should choices regarding
the types of TLDs included in the initial introduction seek to promote effective
evaluation of:
the feasibility
and utility of different types of new TLDs?
the efficacy of different procedures for launching new TLDs?
different policies under which the TLDs
can be administered in the longer term?
different operational models for the registry and registrar functions?
different institutional structures for the
formulation of registration and operation
policies within the TLD?
other
factors?
BT proposes that ICANN should address the introduction of different types
of new TLDs in a phased programme addressing the various types that are proposed.
Clearly some priorities will need to be selected and this could be based on addressing
the most pressing issues which are currently faced - the need for more diversity
from the existing TLDs.
New TLDs should be launched in a variety of different
ways, but their mission, objectives, functional activities and operations should
comply with derived ICANN policies which reflect 'best practice' in TLD operations
and are compliant with applicable legal obligations in chosen territories of operation.
BT considers that the operational models for registry and registrar operations
should satisfy the criteria of firstly, the operational stability of the Internet
and the Root Server System including the rapid and effective maintenance of address
information, new registrations and changes; secondly enhancing the ease of use for
ordinary users of the internet, whether for personal, consumer or business needs
and thirdly, the introduction of competition into registrar functions.
Q16:
Should any particular goal for, or limit on, the number of TLDs to be included in
the initial introduction be established in advance, or alternatively should the number
included in the initial introduction be guided by the extent to which proposals establish
sound proofs of concept of varied new TLD attributes?
BT supports the view that
the initial introduction should be limited to a few new TLDs, pending assessment
of the effects on the stability of the Internet and operational success and growth
of new domains. There is clearly a need to address a variety of new TLD attributes,
and the initial trial domains could be based on proposals arising from these options.
Q17:
In view of the current competitive conditions, should the promotion of effective
competition in the provision of registration services continue to be a significant
motivation for adding fully open TLDs?
BT considers that the interests of the Internet
user should be paramount and supports the position of the Business and Commercial
constituency in encouraging ICANN to adopt principles which have clear meanings whereby
a gTLD should be intuitively recognisable and reflect how the new TLD will be perceived
by the relevant population of net users. Within this principle there will be scope
for competition and differentiation, which should be managed to control pressures
which may effect the stability of the DNS.
Q18: Should the desire for diverse
vendors of registry services in open TLDs be an important motivation in adding fully
open TLDs?
What will not serve users well would be semantic duplication of functional
TLDs which will cause confusion and frustrate users, and reduce ease of access to
sites within these TLDs. For this reason it would appear better to sustain the uniqueness
of TLDs and their associated registries and encourage competition amongst registrars.
Q19:
Would the introduction of additional undifferentiated TLDs result in increased inter-TLD
confusion among Internet users?
What will not serve users well would be semantic
duplication of functional TLDs which will cause confusion and frustrate users, and
reduce ease of access to sites within these TLDs. For this reason it would appear
better to sustain the uniqueness of TLDs and their associated registries and encourage
competition amongst registrars. What will be unacceptable to businesses will be the
proliferation of competing domains which forces them to register everywhere so increasing
business costs and consumer confusion.
Q20: Taking all the relevant factors into
account, should one or more fully open TLDs be included in the initial introduction?
Yes
but with a clear vision for how this will benefit users in the future. These need
to address two principle issues, firstly providing greater differentiation from existing
domains which provide a 'catchall' and secondly those for which the existing TLDs
do not provide any reasonable home. These new open TLDs should all possess a clear
charter.
Q21: How many?
No more than three to allow for other types of TLD.
Q22:
How effective would other fully open TLDs be in providing effective competition to
.com?
That would depend on their declared objectives of the new TLDs. The introduction
of .bus, .biz , would only generate confusion for users, consumers and businesses,
as all entities from .com would need to register protectively in these areas to obviate
cybersquatting or warehousing in certain territories, and to ensure that there is
no loss of trade by absence from a competing domain. If the new open TLDs were more
specifically defined, .banc, .shop. .ins, etc then migration into these areas by
relevant registrants might avoid these dangers. However registrants will probably
not migrate away from .com in the short to medium term and restructuring it is probably
not a viable option. Companies for which the new TLD name space was relevant
would register in it as well. A more specific TLD would provide some extra name space
and companies with the same name would be able to register in their correct semantic
or functional domain. However whether this is effective competition in its self is
questionable.
Q23: What can be done to maximize the prospect that new fully open
TLDs will be attractive to consumers as alternatives to .com?
The most relevant
option would be to introduce more structure but much investigation should be undertaken
before implementing this. This could mean that companies again have to register multiple
times in different directory structures.
Q24: Would the likelihood of effective
competition with .com be enhanced by making one or more of the single-character .com
domains (which are currently registered to the IANA) available for use as the basis
of a third-level registry (i.e. a registry that took registration of names in the
form of example.e.com or example.1.com)? Should the single-character .com domains
be made available for possible registry usage in conjunction with the initial group
of additional TLDs?
BT considers that this would be the wrong solution because
it would result in a completely meaningless name structure that is impossible for
a user to guess. ICANN needs to release a structured competitor to .com which users
will find easy to use and in which businesses can find a clear home
Q25: Is increasing
the utility of the DNS as a resource-location tool an appropriate goal in the introduction
of new TLDs?
The main policy issue is whether the DNS should be seen as a resource-location
tool or not. If the aim of introducing alphanumeric identifiers fronting IP addresses
was to improve the utility of the DNS in locating resources, then it would be illogical
to discontinue that goal, unless it has proved unworkable. In reality many people
now use an intuitive approach to using the DNS as a resource-location tool, and so
it can only be seen as having achieved some success. Consequently BT supports the
view that ICANN should improve its functionality of the DNS as a resource-locator
tool.
Q26: Would the introduction of unrestricted, undifferentiated TLDs run counter
to this goal?
Yes.
Q27: If so, are there ways of accommodating the goal of enhancing
registry-level competition with the goal of enhancing the utility of the DNS?
No.
Although One way might be to add .com1 to compete against .com and enable the DNS
to resolve into either. A company registered in .com1 is automatically registered
in .com (collisions avoided) leading to two operators but a single namespace. The
DNS resolves between the two spaces automatically. This could be extended into other
.comX. Another way would be to split .com between operators alphabetically…a-m.com
plus n-z.com. However double resolution is unsatisfactory and not to be condoned
because it would mean changes to all resolvers and is basically unworkable. The tender
process should provide registry competition much like the competition process for
running a national lottery for a fixed period of time.
Q28: Is the concept of
TLD "charters" helpful in promoting the appropriate evolution of the DNS?
It is
an option but not the only solution. The key issue has to be ease of resource location
for users and some differentiated framework. Because the charter cannot predict the
future use an innovative operator of a TLD could conceive in this dynamic environment,
an innovative operator should be able to request a change to the charter if required,
which could be considered appropriately and sanctioned by ICANN.
Q29: Are the first
three principles outlined in the second additional consensus point of WG-C's 17 April
2000 supplemental report (quoted below) appropriate criteria for selecting TLDs to
be introduced in the first group?
"1. Meaning: An application for a TLD should
explain the significance of the proposed TLD
string,
and how the applicant contemplates that the new TLD will be perceived by the relevant
population of net users. The application may contemplate that
the proposed TLD string will
have its primary semantic
meaning in a language other than English.
"2. Enforcement:
An application for a TLD should explain the mechanism for charter
enforcement where relevant and desired.
"3. Differentiation: The selection of a TLD string should not confuse net users,
and so TLDs
should be clearly differentiated by the
string and/or by the marketing and functionality
associated
with the string."
BT supports these principles for the adoption of TLDs.
Q30:
Do those principles preclude the introduction of any new fully open TLDs?
No but
they could be equally applicable to fully open TLDs
Q31: What types of TLDs should
be included in the first group of additional TLDs to best test the concept of chartered
TLDs?
The selection of trial chartered TLDs could be those which might service
interest groups, and be drawn from commercial and non-commercial entities, reflecting
service providers such as banks or insurance companies, professional groups such
as engineers, architects, lawyers etc and social entities such as ethnicity groups,
or content providers.
Q32: Should chartered TLDs be introduced according to a
pre-defined system, or should proposals be evaluated on an individualized basis?
The
optimum solution might be a combination of both, whereby there can be individualised
TLDs within some semantic framework, encouraging uniqueness to avoid user confusion.
Q33:
If charter proposals are evaluated on an individualized basis, should any steps should
be taken to promote stable and orderly evolution of the DNS overall?
This depends
on the evaluation process and how TLD are evaluated, but a system of measurement
and verification should be applied. In addition to receiving direct applications
for TLD from registrars, ICANN could raise charters themselves and request applications
from registries to run these TLDs. These charters should be raised after open debate/requests/comments
for new TLD, on the basis of recommendations from a TLD evaluation committee.
Q34:
Has the inventory of useful and available domain names reached an unacceptably low
level?
While evidence has been quoted by some to suggest that this is the case
in one top level domain, it is not the case generally.
Q35: Assuming it is important
to increase the inventory of available domain names, should that be done by adding
TLDs that are not differentiated from the present ones?
No. The domains should
be differentiated. In addition there have to be some lessons learned from the operation
of the first generation of open TLDs. A semantic or functional framework for
sub-domains could provide a significant resolution of the current limitations for
existing gTLDs.
Q36: Should the formulation of policies for limited-purpose TLDs
be delegated to sponsoring organisations? In all cases or only in some?
Delegation
is an inherent part of DNS and is currently used for subspaces of existing TLDs.
It would be appropriate for any delegation to be consistent with the assigned charter
but ICANN should require all applicants for TLDs to indicate how they propose to
manage the available address space within the domain in an economical and effective
manner for users of the domain.
Q37: What measures should be employed to encourage
or require that a sponsoring organization is appropriately representative of the
TLD's intended stakeholders?
BT supports the position that ICANN should adopt appropriate
measures to avoid future land-grabs and compromises of stakeholders' rights as formerly
happened with trademarks. Stakeholders should be representative or be acting on behalf
of these stakeholders and measures should include a grievance resolution procedure
and a clear charter. However ICANN should also address how this can be applied to
the existing operators of the domains described by .com, .net and .org and
what needs to be done to resolve any inconsistencies in the criteria. This issue
could constrain potential creators of new TLDs who wish to compete commercially and
should be equitably applied.